USPTO Patents Application 10770421

IN THE CLAIMS

Please amend Claims 13, 14, 16, 18-24, 37, 38, 40, 42-48, 50 and 52, to read as

follows.

1-12. (Canceled)

13. (Currently Amended) An apparatus for determining the location of a boundary
between a speech containing portion and a background noise containing portion in an input
speech signal, the apparatus comprising:

means for receiving the input signal;

means for processing the received signal to generate an energy signal indicative of the
local energy within the received signal;

speech detection means operable to process said tfie received signal and to identify
when speech is present in the received signal;

means for determining the likelihood that said the boundary is located at each of a
plurality of possible locations within said the energy signal; and

means for determining the location of said the boundary using said the likelihoods
determined for each of said ttie possible locations,

wherein said likelihood determining means is operable restricted to determine said the
likeHhoods in the received signal onlv when said speech detecting means detects speech within
the received signal.

-2-

14. (Currently Amended) An apparatus according to claim 13, wherein said likelihood
determining means is operable to determine the likelihood that said tiie boimdary is located at
each of said tiie possible locations by: (i) comparing a portion of the energy signal on one side of
the current location with a model representative of the energy in background noise; (ii)
comparing tiic a portion of the energy signal on the other side of the current location with a
model representative of the energy within speech; and (iii) combining the results of said tfie
comparisons to determine a likelihood for the current possible location.

15. (Canceled)

16. (Currently Amended) An apparatus according to claim 13, further comprising
means for filtering said the energy signal to remove energy variations which have a frequency
below a predetermined frequency.

17. (Original) An apparatus according to claim 16, wherein said filter means is
operable to filter out energy variations below IHz.

18. (Currently Amended) An apparatus according to claim 13, wherein said
processing means is operable to divide the input speech signal into a number of successive time
frames and to determine the energy of the input signal in each of said tfie time frames to generate
a discrete energy signal.

-3-

19. (Currently Amended) An apparatus according to claim 16, wherein said filter
means is operable to output a number of discrete samples representing said the filtered energy
signal.

20. (Currently Amended) An apparatus according to claim 19, wherein said
likelihood determining means is operable to determine said Ae likelihood for each of said tiie
discrete filtered energy values.

21. (Currently Amended) An apparatus according to claim 13, wherein said the
boundary is at the beginning or at the end of a speech containing portion of said the received
signal.

22. (Currently Amended) An apparatus according to claim 14, wherein said the
models are statistical models.

23. (Currently Amended) An apparatus according to claim 22, wherein said the
models are based on Laplacian statistics.

24. (Currently Amended) An apparatus according to claim 22, wherein said speech
mod e l is an the models are auto-regressive model models.

25-36. (Canceled)

-4-

37. (Currently Amended) A method of determining the location of a boundary
between a speech containing portion and a background noise containing portion in an input
speech signal, the method comprising the following steps [[of]]:

receiving the input signal;

processing the received signal to generate an energy signal indicative of the local
energy within the received signal;

a speech detection step which processes said tiie received signal and identifies when
speech is present in the received signal;

determining the likelihood that said the boundary is located at each of a plurality of
possible locations within said tiie energy signal; and

determining the location of said tiie boundary using said tiie likelihoods determined for
each of said the possible locations,

wherein said likelihood determining step dctcnnines said is restricted to determine the
likelihoods in the received signal onlv when said speech detecting step detects speech within the
received signal.

38. (Currently Amended) A method according to claim 37, wherein said likelihood
determining step determines the likelihood that said ttie boundary is located at each of said the
possible locations by: (i) comparing a portion of the energy signal on one side of the current
location with a model representative of the energy in background noise; (ii) comparing tiic a
portion of the energy signal on the other side of the current location with a model representative

-5-

of the energy within speech; and (iii) combining the results of said fee comparisons to determine
a UkeUhood for the current possible location.

39. (Canceled)

40. (Currently Amended) A method according to claim 37, further comprising tiic a
step of filtering said the energy signal to remove energy variations which have a frequency below
a predetermined frequency.

41. (Original) A method according to claim 40, wherein said filtering step filters out
energy variations below IHz,

42. (Currently Amended) A method according to claim 37, wherein said processing
step divides the input speech signal into a number of successive time frames and determines the
energy of the input signal in each of sard tiie time frames to generate a discrete energy signal.

43. (Currently Amended) A method according to claim 40, wherein said filtering step
outputs a number of discrete samples representing said tfie filtered energy signal.

44. (Currently Amended) A method according to claim 43, wherein said likelihood
determining step determines said the likeUhood for each of said the discrete filtered energy
values.

-6-

45. (Currently Amended) A method according to claim 37, wherein said the boundary
is at the beginning or at the end of a speech containing portion of the received signal.

46. (Currently Amended) A method according to claim 38, wherein said the models
are statistical models.

47. (Currently Amended) A method according to claim 46, wherein said the models
are based on Laplacian statistics.

48. (Currently Amended) A method according to claim 46, wherein said speech
model is an the models are auto-regressive mod e l models ,

49. (Canceled)

50. (Currently Amended) A computer readable medium storing computer executable
process steps for controlling a processor to implement a method of detecting speech wtttt within
an input signal, the process steps comprising the steps of:

receiving the input signal;

processing the received signal to generate an energy signal indicative of the local
energy within the received signal;

processing said the received signal to identify when speech is present in the received

signal;

determining the likelihood that said tfie boundary is located at each of a plurality of
possible locations within said the energy signal; and

determining the location of said ttie boundary using said the likelihoods determined for
each of said the possible locations,

wherein said likelihood determining d c tcmiin c s said step is restricted to determine the
likelihoods in the received signal only when speech is detected within the received signal

51. (Canceled)

52. (Currently Amended) Computer executable process steps for controlling a
processor to implement a method of detecting the presence of speech wife within an input signal,
the process steps comprising the steps of:

receiving the input signal;

processing the received signal to generate an energy signal indicative of the local
energy within the received signal;

processing said the received signal to identify when speech is present in the received

signal;

determining the likelihood that said die boundary is located at each of a plurality of
possible locations within said the energy signal; and

determining the location of said tiie boundary using said the likelihoods determined for
each of said the possible locations,

-8-

wherein said likelihood determining dctcmiincs said step is restricted to determine the
likelihoods in the received signal only when speech is detected within the received signal.

-9-